Patents Assigned to Beijing Baidu Netcom Science and Technology Co., Ltd
  • Patent number: 12026966
    Abstract: A method for recognizing a dialogue intention, an electronic device and a storage medium are provided. The solution includes: obtaining a sample sentence with an intention category to be predicted and a plurality of supporting sentences each labeled with an intention category; and inputting the sample sentence and the plurality of supporting sentences into a trained intention recognition model, in which the trained intention recognition model is configured to generate a sample sentence vector corresponding to the sample sentence and category vectors corresponding to the plurality of supporting sentences based on the sample sentence and the plurality of supporting sentences, calculate matching degrees between the sample sentence vector and the category vectors, and obtain a predicted intention category of the sample sentence based on the matching degrees.
    Type: Grant
    Filed: September 15, 2021
    Date of Patent: July 2, 2024
    Assignee: BEIJING BAIDU NETCOM SCIENCE AND TECHNOLOGY CO., LTD.
    Inventors: Hongjie Guo, Hongyang Zhang, Zhenyu Jiao, Shuqi Sun, Yue Chang, Tingting Li
  • Patent number: 12026937
    Abstract: Methods, apparatuses, electronic devices, and mediums related to classification models are provided. The method may include: obtaining a set of positive samples and a set of negative samples, a sample in the set of positive samples including a pair of images from a first image set annotated as similar with each other, and a sample in the set of negative samples including a pair of images annotated as dissimilar with each other from the first image set; training a first sub-network based on the set of positive samples and the set of negative samples; constructing a classification model including a second sub-network, where initial weights of the second sub-network are weights of corresponding nodes of the trained first sub-network; and training the classification model using a second image set that includes at least one image that is common with at least one image in the first image set.
    Type: Grant
    Filed: November 19, 2021
    Date of Patent: July 2, 2024
    Assignee: BEIJING BAIDU NETCOM SCIENCE TECHNOLOGY CO., LTD.
    Inventor: Yuting Du
  • Patent number: 12025459
    Abstract: Provided are methods and apparatuses for navigation guidance and establishing a three-dimensional real scene model, a device and a medium, which relate to the field of artificial intelligence and, in particular, intelligent transportation technologies.
    Type: Grant
    Filed: August 25, 2021
    Date of Patent: July 2, 2024
    Assignee: BEIJING BAIDU NETCOM SCIENCE AND TECHNOLOGY CO., LTD.
    Inventors: Pingnan Chang, Ye Sun, Chen Chen, Jia Zhou, Shanshan Liu, Jin Tian, Yao Xiao
  • Patent number: 12026547
    Abstract: A method and a system for arranging business process, a computing device, and a non-transitory computer readable storage medium are provided. The method includes: receiving an application module and a business process rule that are transmitted by a user terminal, wherein the application module is configured to indicate a processing logic in a link in a business process, and wherein the business process rule is configured to indicate a processing rule of the application module; determining a target edge device corresponding to the application module; and transmitting the application module and the business process rule to the target edge device for the target edge device to execute the application module according to the business process rule.
    Type: Grant
    Filed: June 16, 2021
    Date of Patent: July 2, 2024
    Assignee: BEIJING BAIDU NETCOM SCIENCE AND TECHNOLOGY CO., LTD.
    Inventors: Jun Liang, Leding Li
  • Patent number: 12025443
    Abstract: The present application discloses a method and apparatus for identifying an updated road, a device and a computer storage medium, and relates to the field of big data technologies. A specific implementation solution is as follows: comparing a road area extracted based on the latest satellite image with a road area extracted based on a historical satellite image, to obtain a candidate updated road; mapping the candidate updated road into road network data according to a coordinate position of the candidate updated road; acquiring a user trajectory set corresponding to the candidate updated road within a recent preset period; and identifying, based on a matching result between the user trajectory set and the road network data, whether the candidate updated road is an actual updated road. Updated roads can be more accurately identified through the method according to the present application.
    Type: Grant
    Filed: August 24, 2021
    Date of Patent: July 2, 2024
    Assignee: BEIJING BAIDU NETCOM SCIENCE AND TECHNOLOGY CO., LTD.
    Inventors: Jizhou Huang, Deguo Xia, Haifeng Wang
  • Patent number: 12026864
    Abstract: Provided are a method and apparatus for detecting a status of a lens, an electronic device and a medium. A specific implementation includes: determining current feature points included in a current image according to the current image captured at a current moment by a to-be-detected lens of a target driving device; determining historical feature points included in a historical image according to the historical image captured at a historical moment by the to-be-detected lens; and matching the current feature points with the historical feature points and determining a status detection result of the to-be-detected lens according to a matching result. Therefore, an effect of determining a status of the to-be-detected lens of the target driving device is achieved and the problem of a potential safety hazard of the target driving device due to an abnormal status of the to-be-detected lens can be avoided.
    Type: Grant
    Filed: September 20, 2021
    Date of Patent: July 2, 2024
    Assignee: Beijing Baidu Netcom Science Technology Co., Ltd.
    Inventor: Lihua Xu
  • Patent number: 12026230
    Abstract: Technical solutions relate to the field of artificial intelligence such as deep learning, computer vision and intelligent imaging. A method may includes during training of a one-stage object detecting model, acquiring values of a loss function corresponding to feature maps at different scales respectively in the case that classification loss calculation is required, and the loss function is a focal loss function; and determining a final value of the loss function according to the acquired values of the loss function, and training the one-stage object detecting model according to the final value of the loss function.
    Type: Grant
    Filed: December 20, 2021
    Date of Patent: July 2, 2024
    Assignee: BEIJING BAIDU NETCOM SCIENCE TECHNOLOGY CO., LTD.
    Inventors: Binghong Wu, Yehui Yang, DaLu Yang, Yanwu Xu, Lei Wang, Qian Li
  • Publication number: 20240212703
    Abstract: A method of processing audio data, which relates to a field of speech synthesis technology. The method includes: decomposing original audio data to obtain voice audio data and background audio data; performing electroacoustic processing on the voice audio data to obtain electroacoustic voice data; and combining the electroacoustic voice data and the background audio data to obtain target audio data. An electronic device and a storage medium are further provided.
    Type: Application
    Filed: March 22, 2022
    Publication date: June 27, 2024
    Applicant: BEIJING BAIDU NETCOM SCIENCE TECHNOLOGY CO., LTD.
    Inventors: Yipeng WANG, Yunfeng LIU
  • Publication number: 20240211126
    Abstract: A human-machine interaction method, includes: determining a target controlled object corresponding to a first gesture in a first area of a touch pad in response to detecting the first gesture in the first area; determining a target control mode corresponding to a second gesture in a second area of the touch pad in response to detecting the second gesture in the second area; and controlling the target controlled object based on the target control mode.
    Type: Application
    Filed: December 2, 2021
    Publication date: June 27, 2024
    Applicant: BEIJING BAIDU NETCOM SCIENCE TECHNOLOGY CO., LTD.
    Inventors: Houqiang Zhu, Pengfei Wei
  • Publication number: 20240214637
    Abstract: A method of pushing a video, an electronic device and a storage medium are provided, which relate to a field of data processing technology, in particular to big data technology. The method includes: acquiring first high-quality videos; acquiring posterior data obtained when each first high-quality video is distributed for a first preset number of times as a label for the video; acquiring a first preset number of users earliest interacting with each first high-quality video; generating training samples using the videos as video-side features and the users as user-side features to train at least one ranking model; acquiring a first target user; acquiring one or more candidate videos corresponding to each opinion leader to form a first video set; and pushing, after ranking for the first video set using the at least one trained ranking model, a top second preset number of videos to the first target user.
    Type: Application
    Filed: March 15, 2022
    Publication date: June 27, 2024
    Applicant: BEIJING BAIDU NETCOM SCIENCE TECHNOLOGY CO., LTD.
    Inventors: Tiantian HE, Yuanyuan HU
  • Publication number: 20240211609
    Abstract: A method of protecting a model, which relates to a field of computer, a field of artificial intelligence, and may be applied to an AI model protection scenarios. The method includes: generating a WASM file for providing a runtime environment for a target model, the WASM file containing a corresponding model inference algorithm and security verification algorithm, wherein the security verification algorithm is configured to perform at least one security verification operation to protect the target model, the at least one security verification operation is selected from: a verification of a host environment; a verification of an integrity of the WASM file; a verification of an integrity of the model file generated corresponding to an original model file of the target model; a timeout verification of a specified inference process during a model inference process; or a timeout verification of an entire inference process during the model inference process.
    Type: Application
    Filed: March 22, 2022
    Publication date: June 27, 2024
    Applicant: BEIJING BAIDU NETCOM SCIENCE TECHNOLOGY CO., LTD.
    Inventors: Shuangyan YUE, Zhongkai FAN
  • Patent number: 12019592
    Abstract: A file moving method, an electronic device, and a computer-readable storage medium are provided. An implementation includes: in response to receiving a file moving request, determining a source path and a destination path corresponding to a first moving operation corresponding to the request; determining whether the source path and the destination path corresponding to the first moving operation exist; in response to determining that both the source path and the destination path corresponding to the first moving operation exist, comparing at least one of the source path and the destination path corresponding to the first moving operation with a corresponding path of a source path and a destination path corresponding to the second moving operation; and determining, based on a result of the comparing, whether the first moving operation is to be performed.
    Type: Grant
    Filed: February 23, 2022
    Date of Patent: June 25, 2024
    Assignee: BEIJING BAIDU NETCOM SCIENCE TECHNOLOGY CO., LTD.
    Inventor: Yongqiang Yang
  • Patent number: 12020467
    Abstract: The present disclosure provides a method for optimizing a tag of a point of interest s(POI). The method includes: obtaining first portrait feature data of each POI in a plurality of POIs and second portrait feature data of each tag in a plurality of marked tags corresponding to the plurality of POIs; mapping the first portrait feature data of each POI and the second portrait feature data of each tag to a metric space to obtain a first feature vector of each POI and a second feature vector of each tag; and optimizing at least one marked tag corresponding to a target POI based on a vector similarity between a first feature vector of the target POI and a second feature vector of at least one tag. The present disclosure provides an apparatus for optimizing a tag of a POI, an electronic device and a computer readable medium.
    Type: Grant
    Filed: September 29, 2020
    Date of Patent: June 25, 2024
    Assignee: BEIJING BAIDU NETCOM SCIENCE AND TECHNOLOGY CO., LTD.
    Inventors: Jingbo Zhou, Renjun Hu, Airong Jiang, Jianguo Duan, Hui Xiong
  • Patent number: 12019990
    Abstract: The present application discloses a text processing method and device based on natural language processing and a knowledge graph, and relates to the in-depth field of artificial intelligence technology. A specific implementation is: an electronic device uses a joint learning model to obtain a semantic representation, which is obtained by the joint learning model by combining knowledge graph representation learning and natural language representation learning, it combines a knowledge graph representation learning and a natural language representation learning, compared to using only the knowledge graph representation learning or the natural language representation learning to learn semantic representation of a prediction object, factors considered by the joint learning model are more in quantity and comprehensiveness, so accuracy of semantic representation can be improved, and thus accuracy of text processing can be improved.
    Type: Grant
    Filed: December 16, 2020
    Date of Patent: June 25, 2024
    Assignee: BEIJING BAIDU NETCOM SCIENCE TECHNOLOGY CO., LTD.
    Inventors: Haifeng Wang, Wenbin Jiang, Yajuan Lv, Yong Zhu, Hua Wu
  • Patent number: 12014170
    Abstract: A mini program batch processing method and apparatus, an electronic device, and a readable storage medium are provided. A method includes: establishing a batch package submission creation event; creating, in response to the batch package submission creation event, a batch package submission task to generate a package submission work queue comprising a plurality of mini program package submission tasks; extracting a first quantity of mini program package submission tasks from the package submission work queue; creating a virtual mini program package for each of the first quantity of mini program package submission tasks; and updating the virtual mini program package to a mini program online package and automatically releasing the mini program online package.
    Type: Grant
    Filed: August 9, 2021
    Date of Patent: June 18, 2024
    Assignee: Beijing Baidu Netcom Science and Technology Co., Ltd.
    Inventors: Hao Zhang, Yuanxu Liu, Ke Tao
  • Patent number: 12008820
    Abstract: The present disclosure provides a method of matching lane line data, which relates to a field of automatic driving, and in particular to a field of locating an autonomous vehicle. The method of matching lane line data includes: obtaining a first data indicating sensed lane lines and a second data indicating map lane lines respectively; constructing a correlation weight matrix for the first data and the second data, based on differences between the sensed lane lines and the map lane lines; and matching the first data with the second data according to the correlation weight matrix, so as to obtain a third data having an optimal match with the first data from the second data. The present disclosure further discloses an apparatus of matching lane line data, a device, a storage medium and a computer program product.
    Type: Grant
    Filed: June 18, 2021
    Date of Patent: June 11, 2024
    Assignee: BEIJING BAIDU NETCOM SCIENCE TECHNOLOGY CO., LTD.
    Inventors: Pengbin Yang, Yang Yang, Jianxu Zhang, Wenlong Chen
  • Patent number: 12007965
    Abstract: The disclosure provides a method for deduplicating entity nodes in a graph database. The method includes: obtaining a set of entity nodes to be deduplicated from the knowledge graph, in which the set includes a plurality of entity nodes; selecting an untraversed entity node from the set as a target entity node; determining a range located by a node identifier corresponding to the target entity node; determining the target entity node that has appeared in traversed entity nodes according to a deduplicating mode corresponding to the range; and deleting the target entity node from the set.
    Type: Grant
    Filed: May 12, 2022
    Date of Patent: June 11, 2024
    Assignee: BEIJING BAIDU NETCOM SCIENCE TECHNOLOGY CO., LTD.
    Inventors: Yifei Wang, Yang Wang, Yu Wang
  • Publication number: 20240185379
    Abstract: A method and an apparatus for generation a high definition map, a device and a computer storage medium, which relate to automatic driving and deep learning technologies in the field of artificial intelligence technologies, are disclosed. An implementation includes: acquiring point cloud data and front-view image data which are collected respectively by a plurality of collecting devices at a plurality of location points to obtain a sequence of point clouds and a sequence of front-view images; performing registration of the front-view images and the point clouds on the sequence of point clouds and the sequence of front-view images; transforming the sequence of front-view images into a top-view image based on the result of the registration and determining coordinate information of each pixel in the top-view image; and identifying map elements of the top-view image to obtain the high definition map.
    Type: Application
    Filed: November 17, 2021
    Publication date: June 6, 2024
    Applicant: BEIJING BAIDU NETCOM SCIENCE TECHNOLOGY CO., LTD.
    Inventors: Deguo XIA, Jizhou HUANG, Haifeng WANG
  • Patent number: 12001659
    Abstract: Provided are a method of generating a note, an electronic device, and a storage medium, which relate to a field of artificial intelligence, and in particular to fields of information processing and e-book technology. The method includes: displaying, in a process of displaying an e-book reading page, at least one note combination option for a first note generation instruction, in response to the first note generation instruction being detected; determining a second note generation instruction associated with a target note combination option, in response to an operation of selecting the target note combination option from the at least one note combination option; and combining a current character content associated with the first note generation instruction with a target character content associated with the second note generation instruction, so as to generate a combined note content.
    Type: Grant
    Filed: April 13, 2023
    Date of Patent: June 4, 2024
    Assignee: Beijing Baidu Netcom Science Technology Co., Ltd.
    Inventors: Yongzheng Xin, Shuangshuang Cui, Wensi Su, Huibin Zhao
  • Publication number: 20240177469
    Abstract: A method and apparatus for encoding a geographic location region as well as a method and apparatus for establishing an encoding model, which relate to big data and deep learning technologies in the field of artificial intelligence technologies are disclosed. An implementation includes: determining a to-be-encoded geographic location region; acquiring at least one kind of geographic function information and at least one kind of surface-feature distribution information of the geographic location region; and inputting the acquired geographic function information and the acquired surface-feature distribution information into an encoding model, the encoding model performing embedding on the geographic function information and the surface-feature distribution information, and fusing vector representations obtained by the embedding to obtain an encoding result of the geographic location region.
    Type: Application
    Filed: November 17, 2021
    Publication date: May 30, 2024
    Applicant: BEIJING BAIDU NETCOM SCIENCE TECHNOLOGY CO., LTD.
    Inventors: Miao FAN, Jizhou HUANG, Haifeng WANG